As shown in FIG. 8, in order to dehumidify a sample fluid, remove dusts from the sample fluid, keep the flow rate of the sample fluid constant, etc. in a passage from a sample extracting point to an analyzer, a conventional atmospheric pollution measuring apparatus such as a measuring apparatus for a source of gas outbreaks, an ambient air measuring apparatus, a vehicle exhaust gas measuring apparatus or the like is provided with a filter, a changeover valve, a sampling pipe, a dehumidifier, a suction pump, a throttle valve, a flowmeter, etc., and also the respective members are connected to one another through pipes to form a sample flow passage (see Japanese Industrial Standards “JIS B7982-2002”, for example).
There has been also proposed, as shown in FIG. 9, a stack-gas sampling apparatus 101 in which a time required for measuring the concentration of funnel fume (soot and smoke) can be shortened by simultaneously extracting solid components such as ash dust (soot and dust) in the stack-gas and gas components such as sulfur oxides (SOx) and nitrogen oxides (NOx). In this stack-gas sampling apparatus 101, a sampling probe 102 for measuring gas components and a sampling probe 103 for measuring ash dust are arranged in parallel and in proximity to each other by a tap member 104 having openings through which these sampling probes 102 and 103 can be inserted. The insertion length of at least one of the sampling probes 102 and 103 into a gas duct can be adjusted. A filter (not shown) can be installed in or connected to the sampling probe 102 for measuring the gas components, and a dust collector 131 can be also installed in or connected to the sampling probe 103 for measuring the ash dust (see JP-A-2006-226866, for example).
However, the foregoing measuring apparatuses have had some problems. That is, stack-gas from a glass fusion furnace, a boiler of a diesel electric power generator, a metal refining incinerator, a refuse incinerator, an iron and steel fusion furnace, etc. (hereinafter referred to as “high-temperature fusion furnace, etc.”) contains acid mist which is a corrosive substance, such as SO3 mist. Accordingly, when stack-gas is measured by using the foregoing measuring apparatuses, acid mist has been hitherto removed by using a column containing inorganic porous material or granular activated charcoal impregnated with nonvolatile acid in a sample treatment unit. However, it has been defective in long-term stability, and the reason for this has been unclear.